Mesenchymal Stem Cell-Based Therapies for Temporomandibular Joint Repair: A Systematic Review of Preclinical Studies

Temporomandibular disorders (TMDs) are a heterogeneous group of musculoskeletal and neuromuscular conditions involving the temporomandibular joint (TMJ), masticatory muscles, and associated structures. Mesenchymal stromal/stem cells (MSCs) have emerged as a promising therapy for TMJ repair. This systematic review aims to consolidate findings from the preclinical animal studies evaluating MSC-based therapies, including MSCs, their secretome, and extracellular vesicles (EVs), for the treatment of TMJ cartilage/osteochondral defects and osteoarthritis (OA). Following the PRISMA guidelines, PubMed, Embase, Scopus, and Cochrane Library databases were searched for relevant studies. A total of 23 studies involving 125 mice, 149 rats, 470 rabbits, and 74 goats were identified. Compliance with the ARRIVE guidelines was evaluated for quality assessment, while the SYRCLE risk of bias tool was used to assess the risk of bias for the studies. Generally, MSC-based therapies demonstrated efficacy in TMJ repair across animal models of TMJ defects and OA. In most studies, animals treated with MSCs, their derived secretome, or EVs displayed improved morphological, histological, molecular, and behavioral pain outcomes, coupled with positive effects on cellular proliferation, migration, and matrix synthesis, as well as immunomodulation. However, unclear risk in bias and incomplete reporting highlight the need for standardized outcome measurements and reporting in future investigations.


Introduction
The temporomandibular joint (TMJ) is a ginglymoarthrodial (hinging and gliding) joint that connects the mandibular condyle with the temporal articular surface and is one of the most frequently used joints in the human body.Many pathological stimuli such as injury/trauma, joint overload, malocclusion, stress, and psychiatric illness can impact the joint structure and function, potentially leading to temporomandibular disorders (TMDs), a heterogeneous group of musculoskeletal and neuromuscular conditions involving the TMJ, masticatory muscles, and associated structures [1].Among these conditions, osteoarthritis (OA) of the TMJ is an important subtype of TMDs and is characterized by synovial inflammation, disc degeneration, cartilage degradation, and subchondral bone erosion [2].
Patients with TMJ-OA suffer joint pain and dysfunction with a reduced quality of life.Based on clinical and magnetic resonance imaging (MRI) examination, the prevalence of TMJ-OA was found to be approximately 25% in the 20-49 age group, affecting more females than males [3].
There is presently no curative treatment for TMJ-OA, and current management is focused on conservative and non-invasive modalities, including physical therapies, occlusal splints/orthotics, non-steroidal anti-inflammatory drugs (NSAIDs), arthrocentesis, and intra-articular injections of hyaluronic acid (HA) [1,2].Despite providing symptomatic relief to some extent, these treatments are unable to repair and restore the damaged cartilage and subchondral bone.Thus, restoring the integrity and function of the joint tissues, including cartilage and subchondral bone, is critical in halting or reversing the OA progression, where the final treatment option is prosthetic replacement [1].However, repair and regeneration of the joint tissues, in particular the damaged cartilage, have been challenging as the cartilage generally has a poor healing capacity due primarily to it being a poorly vascularized, aneural, and alymphatic load-bearing tissue overlying the supporting subchondral bone.
In recent years, stem cells, particularly mesenchymal stem/stromal cells (MSCs), have emerged as a promising cell source for TMJ repair [4].Residing as multipotent cells within several adult tissues, MSCs have been isolated from diverse tissues, including bone marrow, adipose tissue, skeletal muscle, synovium, dental tissues, peripheral blood, and others [5].They can replicate as undifferentiated cells and have the potential to differentiate into multiple lineages of the mesenchymal tissues, including bone, cartilage, fat, and muscle [5].
MSCs have been reported to be safe and effective for articular cartilage repair of the knee, with the added benefits of lower cost and lesser donor site morbidity [6].Other clinical studies have also reported the efficacy of intra-articular MSC injections to improve pain and functions for knee OA, as recently reviewed [7].In the context of TMJ repair, intra-articular injections of autologous bone marrow nucleated cells [8] or fat-derived stem cells containing MSCs [9] reported improved clinical outcomes in patients with TMJ-OA or derangement.Despite their therapeutic potential, the clinical translation of such an approach using MSCs remains somewhat limited.
Although MSCs were thought to act therapeutically as stem cells via cellular differentiation and replacement, it is now accepted that these cells mediate tissue repair through their paracrine secretion and may be more appropriately termed "medicinal signaling cells" [10].The secreted factors, collectively known as the secretome, are composed of soluble proteins, free nucleic acids and lipids, and extracellular vesicles (EVs).Being a heterogeneous class of lipid membrane vesicles released by cells into the extracellular space, EVs can be broadly classified into different classes, namely, exosomes, microvesicles/microparticles, and apoptotic bodies [11].They are thought to function primarily as intercellular communication vehicles to transfer bioactive cargoes to elicit diverse biological responses in recipient cells, and with MSC-EVs, many of these biological responses culminate in a therapeutic outcome in injured or diseased cells [12].
The current systematic review aims to summarize the results of the existing animal studies that were conducted to evaluate the therapeutic effects of MSCs and their secreted factors, including EVs, in animal models of TMJ cartilage/osteochondral defects and OA.

Purpose
The purpose of this study is to systematically review and evaluate the efficacy of MSCs, their secretome, and their secreted factors, including EVs, for TMJ repair in animal studies.

Systematic Review
This systematic review follows the recommendations of the Cochrane Handbook for Systematic Reviews of Intervention [13] and was conducted according to the Pre-ferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines (Figure 1) [14,15].
The purpose of this study is to systematically review and evaluate the efficacy of MSCs, their secretome, and their secreted factors, including EVs, for TMJ repair in animal studies.

Systematic Review
This systematic review follows the recommendations of the Cochrane Handbook for Systematic Reviews of Intervention [13] and was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines (Figure 1) [14,15].

Search Strategy
The literature search for eligible studies was conducted in PubMed, Embase, Scopus, and the Cochrane Library through 31 July 2023.The medical subject heading (MeSH) terms and keywords used were "temporomandibular joint" or "temporo-mandibular joint" or "temporo mandibular joint" or "craniomandibular" or "cranio-mandibular" or "cranio mandibular" or "TMJ" or "TMD" or "CMD" and "mesenchymal stem cell" or "mesenchymal stromal cell" or "marrow stem cell" or "marrow stromal cell" or "stromal cell" or "MSC" or "secretome" or "exosome" or "extracellular vesicle".

Eligibility Criteria
The PICO framework was used to identify the components of evidence, which are as follows: Population: animals with TMJ condylar defects (cartilage, osteochondral) or TMJ-OA.Intervention: MSC-based therapies comprising MSCs, their secretome, or secreted factors, including EVs for TMJ repair.Comparison: healthy/naive controls, sham, and

Search Strategy
The literature search for eligible studies was conducted in PubMed, Embase, Scopus, and the Cochrane Library through 31 July 2023.The medical subject heading (MeSH) terms and keywords used were "temporomandibular joint" or "temporo-mandibular joint" or "temporo mandibular joint" or "craniomandibular" or "cranio-mandibular" or "cranio mandibular" or "TMJ" or "TMD" or "CMD" and "mesenchymal stem cell" or "mesenchymal stromal cell" or "marrow stem cell" or "marrow stromal cell" or "stromal cell" or "MSC" or "secretome" or "exosome" or "extracellular vesicle".

Eligibility Criteria
The PICO framework was used to identify the components of evidence, which are as follows: Population: animals with TMJ condylar defects (cartilage, osteochondral) or TMJ-OA.Intervention: MSC-based therapies comprising MSCs, their secretome, or secreted factors, including EVs for TMJ repair.Comparison: healthy/naive controls, sham, and other groups (untreated, vehicle, scaffold).Outcome: the histological repair of TMJ and the associated structures as the primary outcome, and the morphological, molecular, and pain improvements as the secondary outcomes.

Study Selection
All search results were loaded into the EndNote 20 reference management software (Clarivate, Philadelphia, PA, USA), and duplicate records were removed.In the first stage, all abstracts and titles were retrieved and screened using the inclusion and exclusion criteria.After the initial screening, the full texts of the articles were further evaluated for eligibility.Three reviewers (Y.J., J.S., and W.D.) independently assessed all titles and abstracts for inclusion.Studies were included if they met the following criteria: in vivo animal studies that used MSCs, their secretome, or EVs for TMJ repair.Language was restricted to English.Studies were excluded if they were not related to TMJ or MSCs, not related to TMJ repair, or presented as case report, conference abstract, review, or purely in vitro study.As illustrated in Figure 1, a PRISMA flow diagram was used to document the study's selection process [16].

Data Extraction
After identifying all the included studies, the following information was recorded: author; year of publication; animal species; age (weeks, months); sample size; gender (male or female); weight (grams or kilograms); animal model (cartilage/osteochondral defect, osteoarthritis); method of induction (surgically, mechanically, or chemically induced); source of MSCs, secretome, or EVs; isolation method; characterization, size distribution, and marker expression; experimental groups; study timepoints; concentration; volume; frequency; delivery; methods of analyses, including gross examination, imaging, histology, immunohistochemistry, molecular analysis, and pain measurement; as well as key in vitro and in vivo therapeutic outcomes.All qualitative and quantitative outcomes of the studies were recorded where available.Fate and biodistribution of MSCs, their secretome, or EVs, and their effects on cell proliferation, migration, and matrix synthesis, as well as immunomodulation, were also evaluated (Figure 2).Attempts were made to contact the corresponding author of the paper when certain details of the study were not reported.
All search results were loaded into the EndNote 20 reference management softw (Clarivate, Philadelphia, PA, USA), and duplicate records were removed.In the first s all abstracts and titles were retrieved and screened using the inclusion and exclusion teria.After the initial screening, the full texts of the articles were further evaluated eligibility.Three reviewers (Y.J., J.S., and W.D.) independently assessed all titles and stracts for inclusion.Studies were included if they met the following criteria: in vivo mal studies that used MSCs, their secretome, or EVs for TMJ repair.Language wa stricted to English.Studies were excluded if they were not related to TMJ or MSCs related to TMJ repair, or presented as case report, conference abstract, review, or pu in vitro study.As illustrated in Figure 1, a PRISMA flow diagram was used to docum the studyʹs selection process [16].

Data Extraction
After identifying all the included studies, the following information was recor author; year of publication; animal species; age (weeks, months); sample size; ge (male or female); weight (grams or kilograms); animal model (cartilage/osteochondra fect, osteoarthritis); method of induction (surgically, mechanically, or chemically duced); source of MSCs, secretome, or EVs; isolation method; characterization, size d bution, and marker expression; experimental groups; study timepoints; concentra volume; frequency; delivery; methods of analyses, including gross examination, imag histology, immunohistochemistry, molecular analysis, and pain measurement; as we key in vitro and in vivo therapeutic outcomes.All qualitative and quantitative outco of the studies were recorded where available.Fate and biodistribution of MSCs, thei cretome, or EVs, and their effects on cell proliferation, migration, and matrix synthes well as immunomodulation, were also evaluated (Figure 2).Attempts were made to tact the corresponding author of the paper when certain details of the study were no ported.

Quality and Risk of Bias Assessments
The quality assessment for each included study was evaluated through their compliance with the ARRIVE (Animal Research: Reporting of In Vivo Experiments) guidelines [17].Twenty items were evaluated for each study.The ratio for each item was calculated by dividing the category score (number of points assigned) by the total score (total points for the item).The quality of the item was assessed as "excellent" for values of the ratio between 0.8 and 1.0, "average" for values of the ratio between 0.5 and 0.79, and "low" for values of the ratio below 0.5.The risk of bias was evaluated using the Systematic Review Center for Laboratory Animal Experimentation (SYRCLE) risk of bias assessment tool for animal studies [18] that assessed selection bias, performance bias, detection bias, attrition bias, reporting bias, and other biases.The parameters assessed include sequence generation, baseline characteristics, allocation concealment, random housing, blinding of investigators, random outcome assessment, blinding of outcome assessor for the different analyses, incomplete outcome data, selective outcome reporting, and other sources of bias.
All included studies were assessed as having high, unclear, or low risk of bias for the different parameters.This assessment was performed by first author (Y.J.) and independently checked by the second and third authors (J.S. and W.D.).Any discrepancy was resolved by discussion with the fourth author (K.Y.W.T.) and the senior author (W.S.T.).

Search Process and Study Selection
The initial database search retrieved 2340 articles, which included 682 in PubMed, 1398 in Embase, 241 in Scopus, and 19 in Cochrane Library.A total of 749 duplicates were removed, and 1560 studies were excluded after screening the title and abstract.After reading the full texts of 31 articles that might meet the inclusion criteria, 6 articles on age-related degeneration, rheumatoid arthritis, osteoporosis, TMJ condylar resorption, growth, and replacement, as well as 2 articles of other languages, were excluded.A PRISMA flow chart showing the study-selection process is presented in Figure 1.A total of 23 studies that fulfilled the inclusion criteria were included in this systematic review .They were published between 2011 and 2022 and more frequently (>50% of papers) between 2020 and 2022 (Figure 3).These 23 included studies were divided according to the nature of the assessed TMDs to enable reliable and equivalent comparisons between studies.These included 11 studies that evaluated MSC-based therapies for the repair of cartilage/osteochondral defects [19][20][21][22][23][24][25][26][27][28][29] and another 12 studies that evaluated MSC-based therapies for TMJ-OA treatment [30][31][32][33][34][35][36][37][38][39][40][41].The outcomes of the studies were largely evaluated qualitatively as there were insufficient quantitative data across the studies for pooling of data quantitatively.
included studies were assessed as having high, unclear, or low risk of bias for the diff parameters.This assessment was performed by first author (Y.J.) and independ checked by the second and third authors (J.S. and W.D.).Any discrepancy was res by discussion with the fourth author (K.Y.W.T.) and the senior author (W.S.T.).

Search Process and Study Selection
The initial database search retrieved 2340 articles, which included 682 in Pub 1398 in Embase, 241 in Scopus, and 19 in Cochrane Library.A total of 749 duplicate removed, and 1560 studies were excluded after screening the title and abstract.After ing the full texts of 31 articles that might meet the inclusion criteria, 6 articles o related degeneration, rheumatoid arthritis, osteoporosis, TMJ condylar resor growth, and replacement, as well as 2 articles of other languages, were exclud PRISMA flow chart showing the study-selection process is presented in Figure 1.A of 23 studies that fulfilled the inclusion criteria were included in this systematic r .They were published between 2011 and 2022 and more frequently (>50% pers) between 2020 and 2022 (Figure 3).These 23 included studies were divided acco to the nature of the assessed TMDs to enable reliable and equivalent comparisons be studies.These included 11 studies that evaluated MSC-based therapies for the rep cartilage/osteochondral defects [19][20][21][22][23][24][25][26][27][28][29] and another 12 studies that evaluated MSCtherapies for TMJ-OA treatment [30][31][32][33][34][35][36][37][38][39][40][41].The outcomes of the studies were largely ated qualitatively as there were insufficient quantitative data across the studies for p of data quantitatively.

Basic Characteristics of the Included Studies
The details of all studies are described in Tables 1-6.Table 1 details the animal species, age, sample size, gender, weight, animal model used, and method of induction.Table 2 details the source of MSCs, secretome, or EVs; isolation method; characterization; size distribution; and marker expression.Tables 3 and 4 detail the comparison groups; study timepoints; concentration of MSCs, secretome, or EVs; volume; frequency; delivery; and methods of analysis, including gross examination, imaging, histology, immunohistochemistry, molecular analysis, and pain measurement.Tables 5 and 6 summarize the key in vitro and in vivo outcomes reported in each paper.Yu, 2021 [25] The bi-layered PLGA-g-PCL scaffold with segregated hydrophilicity-hydrophobicity facilitated chondrogenic differentiation of BMSCs toward top fibrocartilage layer and bottom hyaline cartilage layer.
Wang, 2021 [26] The HA-SH/Col I blend hydrogel-BCP ceramic bi-layered scaffold enhanced proliferation and matrix synthesis of rabbit BMSCs and chondrocytes, as well as osteogenic differentiation of BMSCs.
Rabbit BMSCs/chondrocytes-loaded bi-layer scaffold could effectively promote the regeneration of both fibrocartilage and subchondral bone.

Guastaldi, 2021 [27] ND
The MSCs + hydrogel + biosilica was effective in promoting TMJ condylar cartilage regeneration, as evidenced by intact articular surfaces, maturation, and distribution of chondrocytes along the condyle.Lu, 2015 [31] SDF-1 and RANTES were significantly increased in the UAC cartilage compared to the control cartilage.Migration of BMSCs was enhanced when cocultured with a UAC TMJ condyle and was attenuated in the presence of AMD3100 (CXCR4 antagonist) or BX471 (CCR1 antagonist).
BMSC injections improved cartilage repair and subchondral bone restoration in TMJ-OA mice induced by UAC.The locally injected BMSCs were found to implant and differentiate into chondrocytes in OA cartilage.These effects of BMSCs were inhibited by AMD3100 and BX471.
Zaki, 2017 [32] ND Rabbit BMSCs could safely and effectively repair degenerative changes of rabbit TMJs with bovine collagen II-induced arthritis.
Zhang, 2017 [33] Fluid flow shear stress (FFSS) stimulation induced cell death of superficial and deep zone chondrocytes.Genes associated with chondrocyte hypertrophy and fibrosis were upregulated in deep zone chondrocytes with FFSS stimulation.
BMSCs rescued the damaged cartilage by increasing matrix production and scavenging activity.
Kim, 2019 [34] Human UCMSC lines isolated from different donors showed comparable proliferation ability but varying in vitro capacities for chondrogenesis and expression of marker genes for growth factors and ECM compared to that of BMSCs.
Human UCMSCs exerted anti-inflammatory effects and promoted cartilage regeneration in a rabbit model of TMJ-OA.Medium dose of MSCs was most effective in regenerating the articular cartilage with the highest gene expression levels of growth factors.
Human ESC-MSC exosomes suppressed pain and inflammation and reduced cell apoptosis and matrix degradation while enhancing cell proliferation and matrix synthesis to promote overall TMJ repair and regeneration.
Ogasawara, 2020 [36] LC-MS/MS analysis identified several factors present in SHED-CM that could be involved in processes such as anti-fibrosis, anti-apoptosis, anti-inflammation, proliferation, differentiation, and migration of chondrocytes.
SHED-CM contained multiple therapeutic factors with the potential to promote the regeneration and repair of mechanical-stress-induced mouse TMJ-OA.
Cui, 2020 [37] DPSCs downregulated the expression of MMP3 and MMP13 in fibroblast-like synoviocytes by suppressing STAT1 activation under the inflammatory condition.
DPSC local injection relieved pain, suppressed synovial inflammation, and reduced cartilage degradation and subchondral bone destruction in rats.

K öhnke, 2021 [39] ND
Human adipose-derived MSCs with or without HA were more effective than serum and HA alone in restoring the cartilage thickness in a rabbit TMJ-OA model.

AbuBakr. 2022 [40] ND
BMSC-derived MVs restored damaged condylar structure by suppressing inflammation and matrix degradation in a rat model of TMJ-OA.
Rat BMSC-coated GelMA microspheres endowed with superwettable properties and sustained TGF-β release, can efficiently colonize the bone defect site, release cytokine, and promote cartilage healing.
EVs are commonly measured by their particle and/or protein concentration.Among the studies that investigated the use of EVs, one study reported particle concentration of EVs around 2-4 × 10 9 particles/mL [38], two studies reported protein concentration of about 2 mg/mL [35,40], and one study loaded 5 µg EVs per scaffold for implantation [29].With the use of secretome, Ogasawara et al. [36] adjusted the protein concentration of the conditioned medium to 3 µg/mL with serum-free Dulbecco's Modified Eagle Medium (DMEM).

Delivery
A variety of biomaterial scaffolds/carriers, ranging from platelet derivatives [22,23,28], collagen, gelatin, and hyaluronic acid (HA)-based hydrogels [24,26,27] to silk fibroin [29], poly(lactic-co-glycolic acid) (PLGA), and polycaprolactone (PCL)-based polymeric scaffolds [19,25] have been used to deliver MSCs or their EVs for the repair of cartilage/ osteochondral defects.Biphasic scaffolds were also designed to simulate the specific structures and matrix compositions of the osteochondral tissues [25,26].Of note, Wang and colleagues designed the biomimetic biphasic scaffold comprising self-crosslinking thiolated hyaluronic acid (HA-SH) and collagen I blend hydrogel and biphasic calcium phosphate (BCP) ceramics [26].The blend hydrogel was layered and embedded with chondrocytes and BMSCs to simulate stratifications of the cartilage tissue and BCP ceramics to facilitate bone formation [26].Interestingly, Putnova et al. flushed the defect with the human ADRC solution without using any scaffold [21].

Molecular Outcomes
Six studies performed gene expression analysis of the condylar cartilage specimens [22,30,31,34,35,40].Using a rabbit TMJ osteochondral defect model, Cheng et al. reported significantly upregulated expression of cartilage-specific genes, including Sox9, Col II, and ACAN, in the repaired cartilage following treatment with a cell sheet of BMSCs, and this upregulation was further enhanced when the cells were pre-treated with hydro-static pressure and delivered in a platelet-rich fibrin (PRF) scaffold [22].In the TMJ-OA models, five studies reported a significantly upregulated expression of cartilage-specific genes (Sox9, Col2a1, and ACAN) and downregulated expression of genes associated with inflammation (IL-1β, IL-17, TNF-α, and NF-κB) and matrix degradation (MMP3, MMP13, and ADAMTS5) in animals treated with MSCs or their derived EVs compared with the vehicle-treated and/or untreated animals [30,31,34,35,40].Notably, there was also increased expression of genes such as TGF-β1, IGF-1, and FGF-2, which are growth factors involved in chondrogenic differentiation and cartilage anabolism [34].One study also reported downregulated levels of genes associated with fibrosis (α-SMA), apoptosis (BAX), and pain (Substance P, CGRP, NGF, p75NTR, and TrkA) with EV treatment compared to those of vehicle treatment [35].

Pain Behavioral Outcomes
In three studies, the head withdrawal threshold was measured to evaluate the nociceptive responses of the animals during the course of treatment.In these studies, regardless of the route of administration (intra-articular vs. intravenous), MSCs and their derived secretome, including EVs, were all found to be effective in relieving hyperalgesia of progressive TMJ-OA in mice and rats [35][36][37].However, it is important to note that the secretome and EVs were applied via multiple injections, whereas the cells (i.e., DPSCs) were applied via a single intra-articular injection [35][36][37].

Fate and Biodistribution of MSCs, Their Derived Secretome, or EVs
Among the 18 studies that conducted implantation or intra-articular injections of MSCs, 6 studies performed in vivo tracing of MSCs [19,20,24,31,33,34].In cartilage/ osteochondral defects, transplanted MSCs were found in the newly formed tissues for up to 6 weeks [19] and 12 weeks post-surgery [20] in goats and up to 12 weeks in rabbits [24].On the other hand, two studies reported a rapid decline in transplanted MSCs after 7 days and complete loss of the transplanted cells by 20 days following transplantation in a UAC-induced OA mouse model [31,33].One study, however, reported the presence of transplanted MSCs up to 4 weeks post-injection in a rabbit TMJ-OA model induced by MIA [34].Based on these observations, the persistence of MSCs following administration may vary in cartilage/osteochondral defects and in OA in the different animal models.In contrast to the MSC studies, none of the secretome or EV studies performed in vivo tracing of the secretome factors or EVs [29,35,36,38,40].

Cellular Proliferation, Migration, and Matrix Synthesis
Five studies reported increased proliferation and/or decreased apoptosis or anoikis following treatment with MSCs, their derived secretome, or EVs [23,33,35,36,38].One study reported the upregulated expression of stromal-derived factor (SDF)-1 and regulated upon activation normal T-cell expression and secreted RANTES via osteoarthritic cartilage, and thus a higher capacity of attracting the migration of MSCs toward the degraded cartilage in OA [31].These in vivo findings were supported by the in vitro studies where EV treatment was found to enhance the proliferation and migratory activity of BMSCs and chondrocytes in vitro [29,38].Notably, EVs derived from inflammation-stimulated MSCs significantly enhanced the migration of BMSCs over their unstimulated counterpart [29].

Immunomodulation
Two studies examined the immune cell infiltration during TMJ repair [21,37].Reduced infiltration of CD4+ T cells was observed in the synovial tissue and subchondral bone marrow of OA rats treated with human DPSCs compared to those treated with saline [37].On the other hand, Putnova et al. observed significant immune cell infiltration and inflammatory reaction in both soft and hard TMJ tissues in rabbits treated with human ADRCs despite prior immunosuppression, and these were attributed to the heterogeneous cell populations present in ADRCs [21].In other studies, pro-inflammatory mediators, including IL-1β, TNF-α, IFN-γ, IL-17, and inducible nitric oxide synthase (iNOS) [31,[34][35][36][37]40] were downregulated, while anti-inflammatory cytokines such as TGF-β1 and IL-10 [34] were upregulated in animals treated with MSCs, their derived secretome, or EVs compared with the vehicle-treated and/or untreated animals.In vitro, EVs derived from inflammationstimulated MSCs demonstrated augmented efficacy in suppressing inflammation and repolarizing the THP-1 cells toward the M2-like phenotype with the expression of M2 markers, including CD206, IL-10, and CCL22 [29].These effects of MSC-EVs on the macrophage phenotype and polarization were attributed to the high expression level of miR-27b-3p that regulated macrophage polarization by targeting macrophage colony-stimulating factor (CSF)-1 [29].

Risk of Bias Assessment
The SYRCLE risk of bias tool was used to assess the risk of bias for all the included studies, and the assessment is summarized in Table 9.Most domains had unclear risks due to the lack of information (Table 9).None of the studies reported random sequence generation under selection bias and random housing and blinding under performance bias, and therefore were assigned unclear risk .Most studies reported some form of allocation concealment and baseline characteristics, including the species, strain, sex, or weight of the animals used, and were therefore assigned low risk.Only three studies were assigned a low risk for detection bias secondary to the random selection of animals for outcome assessment [22,24,30], while the rest of the studies were classified as unclear risk [19][20][21]23,[25][26][27][28][29][31][32][33][34][35][36][37][38][39][40][41].Only one study reported blinding of the assessor when performing gross assessment [29], whereas nine studies stated that the assessor was blinded when performing the histological analysis [19,22,29,30,34,35,[37][38][39], and therefore were classified as low risk under detection bias.However, those studies that performed imaging, immunohistochemical, and/or molecular analyses did not report blinding for detection bias and therefore presented unclear risk.Notably, all studies were free of selective outcome reporting and presented no apparent issues that could result in a high risk of bias .

Discussion
The principal finding of systematic review is that MSC-based therapies are efficacious in the treatment of TMJ cartilage/osteochondral defects and OA.This therapeutic efficacy of MSC-based therapies in TMJ repair is demonstrated by cartilage regeneration and subchondral bone restoration, with overall improvements in the morphological, histological, molecular, and behavioral pain outcomes in the studies reviewed .
Of 23 studies reviewed, 18 studies tested MSCs as cell-based therapies [19][20][21][22][23][24][25][26][27][28][30][31][32][33][34]37,39,41], and the other five studies investigated the use of MSC-derived secretome [36] and EVs [29,35,38,40] as "cell-free" therapies.Of note, these five studies were published in recent years from 2019 to 2022 and signal a paradigm shift in the therapeutic mechanism of MSCs in tissue repair from one based on cellular differentiation and replacement to one based on secretion and paracrine signaling.Indeed, the role of paracrine secretion, particularly EVs, in mediating the wide-ranging therapeutic efficacy of MSCs has been increasingly reported [42], and it is now accepted that MSCs exert many if not most of their paracrine effects through the release of EVs [43], membrane vesicles that are secreted by all cell types.
Despite the therapeutic efficacy of MSC-based therapies, several factors/variables have been identified to influence the outcome of intervention with the use of MSCs, their secretome, or EVs for TMJ repair.These factors primarily include the cell source, dosage/concentration, and delivery.Several MSC sources were demonstrated to be efficacious in TMJ repair in both cartilage/osteochondral defects and OA, but the ideal cell source is yet to be identified.Kim et al. demonstrated comparable proliferation and differentiation capabilities of human UCMSCs and BMSCs in vitro; however, this remains to be verified in an animal model in vivo [34].A head-to-head comparison of relevant sources of MSCs, their derived secretome and/or EVs would be required in future studies to determine the most effective source from a TMJ repair perspective.Similarly, a wide range of concentrations of MSCs and their derived EVs was reported.One study evaluated a single intra-articular injection of human UCSCs at varying concentrations in a rabbit model of TMJ-OA and identified the medium concentration of MSCs (2.5 × 10 6 cells/mL) as the optimal dose/concentration for regenerating the articular cartilage [34].However, this dose/concentration remains to be validated using other sources of MSCs.
The frequency of treatment is another factor/variable that should be considered and ideally tested.While some studies recorded the long-term survival of the transplanted BMSCs up to 12 weeks in the osteochondral defects of rabbit and goat [20,24], other studies reported a rapid decline in transplanted cells after 1 week in OA in mice and required multiple weekly injections to constantly supply the BMSCs [31,33].Based on these observations, the persistence of MSCs following implantation or intra-articular injection may vary in cartilage/osteochondral defects and in OA in different animal models.In contrast, none of the secretome or EV studies performed in vivo tracing of the secretome factors or EVs.As such, it would be necessary to determine in future studies the fate and bio-distribution of MSCs, their secretome, and EVs following injection into the joint space to determine their site of action and persistence and to optimize the number of injections for optimal treatment.
Other factors, such as the culture condition and the use of a scaffold, would also influence the treatment outcome in TMJ repair.The culture condition affects the potency of MSCs, their derived secretome, and EVs.For instance, in vitro chondrogenic pre-differentiation or hydrostatic pressure stimulation substantially enhanced the chondrogenic potential of MSCs and, henceforth, their therapeutic efficacy in TMJ repair [22,30].The overexpression of NEL-like protein-1 (NELL-1) or hypoxia-inducible factor (HIF)-1α also reportedly enhanced the efficacy of MSCs in osteochondral repair.In those studies, animals treated with NELL-1 or HIF-1α overexpressing BMSCs displayed enhanced fibrocartilage regeneration and subchondral bone restoration with increased matrix deposition and improved structural organization compared to those treated with native MSCs [19,24].Notably, HIF-1α overexpression was found to enhance the survival of transplanted BMSCs, contributing to improved cartilage repair [24].Alternatively, EVs derived from MSCs stimulated under in the presence of TNF-α and IFN-γ demonstrated augmented efficacy for osteochondral repair compared to those derived from unstimulated counterparts [29].Other studies also reported the delivery of MSCs and their derived EVs in different scaffolds/carriers and demonstrated synergistic enhancements in cartilage and subchondral bone repair [19,20,[22][23][24][25][26][27][28][29][30]39,41].For example, Cheng et al. fabricated a BMSC sheet in the PRF scaffold and showed that the BMSC/PRF construct outperformed the BMSC sheet or PRF alone in cartilage regeneration by enhancing matrix synthesis and promoting the mechanical properties and integration of the neocartilage [22].These studies offer possible strategies to engineer and/or coax the MSCs and their derived EVs to possess the desired therapeutic properties for enhanced TMJ repair.However, it is also important to note that these findings of a particular modification or scaffold being superior to another are all derived from single studies [19,20,[22][23][24][25][26][27][28][29][30]39,41], and the conclusions remain to be verified by more studies.
Although the therapeutic mechanisms underlying MSC-based therapies in TMJ repair remain to be elucidated, it could be summarized from this systematic review of 23 studies that MSCs, including their secretome and EVs, likely exert a multifaceted activity in TMJ repair by enhancing cellular proliferation, migration, and matrix synthesis and modulating the immune reactivity (Figure 5).However, in the 23 studies reviewed, only 3 studies delved into the underlying signaling pathways [35,37,38], and only 2 studies investigated immune cell infiltration during TMJ repair [21,37].These clearly reflect our nascent understanding of the therapeutic mechanisms, particularly the immunomodulatory mechanisms, underlying the effects of MSCs, their derived secretome, or EVs in TMJ repair, which warrants further investigation.The identification of key factors/variables influencing the therapeutic outcomes of MSCs, their secretome, and EVs in TMJ repair and improved understanding of the underlying mechanisms, as summarized in this timely review, will undoubtedly inform the planning and design of future studies.
This systematic review has several limitations.Although most of the studies included in this review appeared to be methodologically sound, the lack of proper reporting rendered many of these studies as having unclear risk in several domains, such as random sequence generation; random housing during the experiment; the blinding of caregivers and investigators; and the blinding of outcome assessors in gross, imaging, and immunohistochemical and molecular analyses, according to the SYRCLE risk of bias assessment tool.This highlighted the importance of detailed reporting by adherence to the ARRIVE guidelines so as to improve the credibility and reliability of studies as having high-quality evidence.Several factors/variables, including the cell source, dosage/concentration, scaffold, and delivery route, were also identified to influence the therapeutic outcome of MSCs, their secretome, or EVs in TMJ repair.However, none of these studies systematically evaluated these variables to determine the optimal cell source, concentration, scaffold, delivery route, and/or frequency of treatment, which would likely require optimization for a specific TMJ cartilage/osteochondral defect or OA condition.The deficiency in quantitative data and heterogeneity in the assessment and reporting of outcomes in many studies also precluded a more rigorous meta-analysis of the studies in this review.Therefore, more efforts to standardize the reporting of methodology and outcomes are required in future studies.This would then allow future studies to have better uniformity and validity, with the possible pooling of results through meta-analysis for more robust data analysis and conclusion.
nascent understanding of the therapeutic particularly the immunomodulatory mechanisms, underlying the effects of MSCs, their derived secretome, or EVs in TMJ repair, which warrants further investigation.The identification of key factors/variables influencing the therapeutic outcomes of MSCs, their secretome, and EVs in TMJ repair and improved understanding of the underlying mechanisms, as summarized in this timely review, will undoubtedly inform the planning and design of future studies.

Figure 1 .
Figure 1.Prisma flow diagram.Search strategy and selection process of the included studies.

Figure 1 .
Figure 1.Prisma flow diagram.Search strategy and selection process of the included studies.

Figure 3 .
Figure 3. Bar chart showing the number of published studies included in the systematic review sorted by the year of publication.

Figure 4 .
Figure 4. Animal species used for models of (A) TMJ cartilage/osteochondral defects and (B) OA.

Figure 4 .
Figure 4. Animal species used for models of (A) TMJ cartilage/osteochondral defects and (B) OA.

Figure 5 .
Figure 5. MSC-based therapies alleviate pathological processes in TMJ cartilage/osteochondral defects and OA through a multifaceted mechanism of enhancing cellular proliferation, migration, and matrix synthesis and modulating the immune reactivity.

Table 1 .
Summary of characteristics of animal models.

Table 2 .
Characterization of MSCs and their derived secretome and EVs in vitro.

Table 3 .
Treatment parameters and analyses for included studies evaluating MSC-based therapies for cartilage/osteochondral defects.

Table 4 .
Treatment parameters and analyses for included studies evaluating the efficacy of MSC-based therapies for OA treatment.

Table 5 .
Summary of key outcomes of included studies evaluating MSC-based therapies for cartilage/osteochondral defects.

Table 6 .
Summary of key outcomes for included studies evaluating MSC-based therapies for OA treatment.

Table 7 .
ARRIVE checklist to evaluate the quality of the included studies for the repair of cartilage/osteochondral defects.

Table 8 .
ARRIVE checklist evaluating the quality of the included studies for OA treatment.